During the evolution of mobile communication systems, different radio access techniques have developed. In recent years one has realized the benefits of being able to utilize parallel systems by one and the same mobile terminal. Therefore, solutions of connecting networks of different access technologies have been developed to allow a user to switch between the different accesses.
Mobile communication systems that include multiple radio access networks employing different radio access technologies are throughout the present disclosure denoted multi-access communication systems. It is assumed that a user may move between different access networks using different access technologies, and the user should at all times be connected to the access network that is considered the “best” one, defined by certain criteria.
Some types of applications require that another party can reach the user, e.g. a telephony application, a video telephony application or an instant messaging application, where the user should be able to receive incoming calls. These applications will put some specific demands on the multi-access system. First of all a mechanism is needed for determining in which access network(s) the user can be found. Secondly, staying reachable within an access network costs resources; battery for the user, and capacity for the access network.
The possibility to make use of multiple access technologies and to be able to choose the best one depending on current needs, environment etc. is currently a hot topic that has been discussed in various publications. The focus in most published material on multi-access communication systems is on how to maintain and move sessions when switching between different access types. Typical examples of such handover and mobility publications are the published US patent applications US2002/0147008 A1 and US2002/0160785 A1.
In order to save battery of the user equipment, it is possible to let the user equipment be reachable by only one access network at a time, even if other access networks are present. In many multi-access communication systems, the user equipment is accessed through this only access network even if another technology would be better suited to handle the call. A set-up procedure has to be run before a switch to a better access technology can be performed. The disadvantage with such an approach is that a significant delay is caused before the best suited access technology can be used.
A set-up procedure e.g. in a cellular access network would at least have to include paging from the network, a reply from the terminal on a random access channel, and some signaling between the two for establishment of communication channels to carry the call and further signaling. Only after these steps, it will be possible to perform the signaling needed to switch to another access type. Detailed procedures differ between cellular technologies but the main principle is the same.
In the published US patent application US2002/0090975 A1, a multi-access communication system comprising two cellular access networks, UMTS and GSM, is disclosed. Since both access networks are coordinated by the same MSC, the MSC has inherent information about the coverage of the two access networks. When a request for a call or other type of communication session is received, a paging is performed in one of the access networks. If the requested communication session is of such a kind that the other access network would be better to handle it, the paging includes a request for switching access network. The set-up of the call is made directly in the most suitable access network. This ensures that the call will not be rejected as a consequence of being set-up over an access network that cannot support the call. The request for switching access network is based on the inherent knowledge about the coverage of the two coordinated access networks.
A problem with many conventional multi-access communication systems is that the resource efficiency for keeping a terminal reachable in more than one access network simultaneously is generally poor. Another problem with the conventional multi-access communication systems is that the time delay for performing a call set-up in any preferred access network generally is too long.